基于 XGBoost 算法的复杂碳酸盐岩岩性测井识别

Identification of complex carbonate lithology by logging based on XGBoost algorithm

碳酸盐岩储层在形成过程中受到多种因素的影响,储层岩性复杂多样,基于测井资料对碳酸盐岩岩性识别具有重要意义。为了解决传统的测井岩性识别方法和机器学习方法对于复杂碳酸盐岩岩性识别准确率不高的问题,以廊固凹陷北部奥陶系碳酸盐岩为例,将XGBoost算法应用于复杂碳酸盐岩岩性识别,并将模型的性能与决策树C4.5算法和支持向量机算法进行对比。结果表明,采用的XGBoost算法的岩性识别模型对研究区碳酸盐岩岩性识别的准确率达到了88.18%,相较于决策树C4.5算法和支持向量机算法准确率均提高了10%左右,且由于XGBoost算法采用多线程和分布式计算的方法,使得训练时间大大缩短。基于XGBoost算法建立的岩性识别模型能够有效地识别复杂碳酸盐岩岩性,为复杂碳酸盐岩岩性的测井识别提供了新的思路。

Carbonate reservoirs are affected by a variety of factors during the formation process,and the reservoir lithology is complex and diverse.Logging data are of great significance for carbonate lithology identification.In order to solve the problem that the traditional logging lithology identification method and traditional machine learning have low recognition accuracy for complex carbonate lithology,taking the Ordovician carbonate rocks in the northern Langgu Depression as an example,based on log data,XGBoost algorithm was appied to lithology identification of complex carbonate rocks,and the performance of the model was comparied with the decision tree C4.5 algorithm and the support vector machine algorithm.The results show that the lithology identification model based on XGBoost algorithm has an accuracy rate of 88.18%for the identification of carbonate lithology in the study area.Compared with decision tree C4.5 and support vector machine,the accuracy rate is increased by about 10%.And the XGBoost algorithm uses multi-threaded and distributed computing methods,the training time is greatly shortened.It shows that the lithology identification model established by XGBoost algorithm can effectively identify complex carbonate lithology and provide a new idea for logging identification of complex carbonate lithology.